Pattern Detection on Glioblastoma’s Waddington Landscape via Generative Adversarial Networks

Abstract

Glioblastoma (GBM) is a highly morbid and lethal disease with poor prognosis. Their emergent properties such as cellular heterogeneity, therapy resistance, and self-renewal are largely attributed to the interactions between a subset of their population known as glioblastoma-derived stem cells (GSCs) and their microenvironment. Identifying causal patterns in the developmental trajectories between GSCs and the mature, well-differentiated GBM phenotypes remains a challenging problem in oncology. The paper presents a blueprint of complex systems approaches to infer attractor dynamics from the single-cell gene expression datasets of pediatric GBM and adult GSCs. These algorithms include Waddington landscape reconstruction, GANs (Generative Adversarial Networks), and fractal dimension analysis. Here I show, a R$\ddot{\mathrm{o}}$ssler-like strange attractor with a fractal dimension of roughly 1.7 emerged in all n = 12 patients’ GAN-reconstructed patterns. The findings suggest a strange attractor may be driving the complex dynamics and adaptive behaviors of GBM in signaling state-space.

KEYWORDS:

• Glioblastoma
• complex systems
• cellular decision-making
• attractors
• patterns
• machine intelligence

Acknowledgements

Thanks to Dr. Mario D’Amico, Dr. Ingo Salzmann, and Dr. Laszlo Kalman of Concordia University (Department of Physics) for mentoring me and help revise/edit the paper. Thanks to Rik Bhattacharja for designing the palette of Figures 1 and 2. Thanks to Dalia Alkayal for designing Figure 3.

Conflict of Interest

There are no competing interests.

Author contribution

AU conceived, conducted the analyses, and wrote the manuscript.

Data availability and codes

The single-cell datasets (expression matrices) are available in the following repositories. Columns are cell barcodes, row names are genes, in all expression (count) matrices.

Pediatric GBM: https://singlecell.broadinstitute.org/single_cell/study/SCP393/single-cell-rna-seq-of-adult-and-pediatric-glioblastoma#study-summary (Neftel et al. 2019).

Expression Matrix (Expression Matrix [log2(TPM/10 + 1)] (Smartseq2): IDHwtGBM.processed.SS2.logTPM.txt.gz.

Adult GSC: https://singlecell.broadinstitute.org/single_cell/study/SCP503 (Richards et al. 2021).

Count matrix: Richards_NatureCancer_GSC_scRNAseq_counts.csv.gz.

scEpath algorithm: https://github.com/sqjin/scEpath (Jin et al. 2018).

GAN lab: https://poloclub.github.io/ganlab/ (Kahng et al. 2019).

FracLac ImageJ Plugin (v2.5) (Box-Count algorithm): http://rsb.info.nih.gov/ij/plugins/fraclac/FLHelp/Introduction.htm